def runPDB2PQR(pdblist,
ff,
outname="",
ph=None,
verbose=False,
selectedExtensions=[],
extensionOptions=utilities.ExtraOptions(),
ph_calc_method=None,
ph_calc_options=None,
clean=False,
neutraln=False,
neutralc=False,
ligand=None,
assign_only=False,
chain=False,
drop_water=False,
debump=True,
opt=True,
typemap=False,
userff=None,
usernames=None,
ffout=None,
commandLine=None,
include_old_header=False):
"""
Run the PDB2PQR Suite
Arguments:
pdblist: The list of objects that was read from the PDB file
given as input (list)
ff: The name of the forcefield (string)
Keyword Arguments:
outname: The name of the desired output file
ph: The desired ph of the system (float)
verbose: When True, script will print information to stdout
When False, no detailed information will be printed (float)
extensions: List of extensions to run
extensionOptions:optionParser like option object that is passed to each object.
propkaOptions:optionParser like option object for propka30.
clean: only return original PDB file in aligned format.
neutraln: Make the N-terminus of this protein neutral
neutralc: Make the C-terminus of this protein neutral
ligand: Calculate the parameters for the ligand in mol2 format at the given path.
assign_only: Only assign charges and radii - do not add atoms, debump, or optimize.
chain: Keep the chain ID in the output PQR file
drop_water: Remove water molecules from output
debump: When 1, debump heavy atoms (int)
opt: When 1, run hydrogen optimization (int)
typemap: Create Typemap output.
userff: The user created forcefield file to use. Overrides ff.
usernames: The user created names file to use. Required if using userff.
ffout: Instead of using the standard canonical naming scheme for residue and atom names, +
use the names from the given forcefield
commandLine: command line used (if any) to launch the program. Included in output header.
include_old_header: Include most of the PDB header in output.
pdb2pka_params: parameters for running pdb2pka.
Returns
header: The PQR file header (string)
lines: The PQR file atoms (list)
missedligandresidues: A list of ligand residue names whose charges could
not be assigned (ligand)
"""
pkaname = ""
lines = []
Lig = None
atomcount = 0 # Count the number of ATOM records in pdb
outroot = utilities.getPQRBaseFileName(outname)
if ph_calc_method == 'propka':
pkaname = outroot + ".propka"
#TODO: What? Shouldn't it be up to propka on how to handle this?
if os.path.isfile(pkaname):
os.remove(pkaname)
start = time.time()
if verbose:
print "Beginning PDB2PQR...\n"
myDefinition = Definition()
if verbose:
print "Parsed Amino Acid definition file."
if drop_water:
# Remove the waters
pdblist_new = []
for record in pdblist:
if isinstance(record, (HETATM, ATOM, SIGATM, SEQADV)):
if record.resName in WAT.water_residue_names:
continue
pdblist_new.append(record)
pdblist = pdblist_new
# Check for the presence of a ligand! This code is taken from pdb2pka/pka.py
if not ligand is None:
from pdb2pka.ligandclean import ligff
myProtein, myDefinition, Lig = ligff.initialize(
myDefinition, ligand, pdblist, verbose)
for atom in myProtein.getAtoms():
if atom.type == "ATOM":
atomcount += 1
else:
myProtein = Protein(pdblist, myDefinition)
if verbose:
print "Created protein object -"
print "\tNumber of residues in protein: %s" % myProtein.numResidues()
print "\tNumber of atoms in protein : %s" % myProtein.numAtoms()
myRoutines = Routines(myProtein, verbose)
for residue in myProtein.getResidues():
multoccupancy = 0
for atom in residue.getAtoms():
if atom.altLoc != "":
multoccupancy = 1
txt = "Warning: multiple occupancies found: %s in %s\n" % (
atom.name, residue)
sys.stderr.write(txt)
if multoccupancy == 1:
myRoutines.warnings.append(
"WARNING: multiple occupancies found in %s,\n" % (residue))
myRoutines.warnings.append(
" at least one of the instances is being ignored.\n")
myRoutines.setTermini(neutraln, neutralc)
myRoutines.updateBonds()
if clean:
header = ""
lines = myProtein.printAtoms(myProtein.getAtoms(), chain)
# Process the extensions
for ext in selectedExtensions:
module = extensions.extDict[ext]
#TODO: figure out a way to do this without crashing...
#tempRoutines = copy.deepcopy(myRoutines)
module.run_extension(myRoutines, outroot, extensionOptions)
if verbose:
print "Total time taken: %.2f seconds\n" % (time.time() - start)
#Be sure to include None for missed ligand residues
return header, lines, None
#remove any future need to convert to lower case
if not ff is None:
ff = ff.lower()
if not ffout is None:
ffout = ffout.lower()
if not assign_only:
# It is OK to process ligands with no ATOM records in the pdb
if atomcount == 0 and Lig != None:
pass
else:
myRoutines.findMissingHeavy()
myRoutines.updateSSbridges()
if debump:
myRoutines.debumpProtein()
if ph_calc_method == 'propka':
myRoutines.runPROPKA(ph, ff, outroot, pkaname, ph_calc_options)
elif ph_calc_method == 'pdb2pka':
myRoutines.runPDB2PKA(ph, ff, pdblist, ligand, verbose,
ph_calc_options)
myRoutines.addHydrogens()
myhydRoutines = hydrogenRoutines(myRoutines)
if debump:
myRoutines.debumpProtein()
if opt:
myhydRoutines.setOptimizeableHydrogens()
myhydRoutines.initializeFullOptimization()
myhydRoutines.optimizeHydrogens()
else:
myhydRoutines.initializeWaterOptimization()
myhydRoutines.optimizeHydrogens()
# Special for GLH/ASH, since both conformations were added
myhydRoutines.cleanup()
else: # Special case for HIS if using assign-only
for residue in myProtein.getResidues():
if isinstance(residue, HIS):
myRoutines.applyPatch("HIP", residue)
myRoutines.setStates()
myForcefield = Forcefield(ff, myDefinition, userff, usernames)
hitlist, misslist = myRoutines.applyForcefield(myForcefield)
ligsuccess = 0
if not ligand is None:
# If this is independent, we can assign charges and radii here
for residue in myProtein.getResidues():
if isinstance(residue, LIG):
templist = []
Lig.make_up2date(residue)
for atom in residue.getAtoms():
atom.ffcharge = Lig.ligand_props[atom.name]["charge"]
atom.radius = Lig.ligand_props[atom.name]["radius"]
if atom in misslist:
misslist.pop(misslist.index(atom))
templist.append(atom)
charge = residue.getCharge()
if abs(charge - int(charge)) > 0.001:
# Ligand parameterization failed
myRoutines.warnings.append(
"WARNING: PDB2PQR could not successfully parameterize\n"
)
myRoutines.warnings.append(
" the desired ligand; it has been left out of\n"
)
myRoutines.warnings.append(" the PQR file.\n")
myRoutines.warnings.append("\n")
# remove the ligand
myProtein.residues.remove(residue)
for myChain in myProtein.chains:
if residue in myChain.residues:
myChain.residues.remove(residue)
else:
ligsuccess = 1
# Mark these atoms as hits
hitlist = hitlist + templist
# Temporary fix; if ligand was successful, pull all ligands from misslist
if ligsuccess:
templist = misslist[:]
for atom in templist:
if isinstance(atom.residue, (Amino, Nucleic)):
continue
misslist.remove(atom)
# Create the Typemap
if typemap:
typemapname = "%s-typemap.html" % outroot
myProtein.createHTMLTypeMap(myDefinition, typemapname)
# Grab the protein charge
reslist, charge = myProtein.getCharge()
# If we want a different naming scheme, use that
if not ffout is None:
scheme = ffout
userff = None # Currently not supported
if scheme != ff:
myNameScheme = Forcefield(scheme, myDefinition, userff)
else:
myNameScheme = myForcefield
myRoutines.applyNameScheme(myNameScheme)
header = printPQRHeader(pdblist,
misslist,
reslist,
charge,
ff,
myRoutines.getWarnings(),
ph_calc_method,
ph,
ffout,
commandLine,
include_old_header=include_old_header)
lines = myProtein.printAtoms(hitlist, chain)
# Determine if any of the atoms in misslist were ligands
missedligandresidues = []
for atom in misslist:
if isinstance(atom.residue, (Amino, Nucleic)):
continue
if atom.resName not in missedligandresidues:
missedligandresidues.append(atom.resName)
# Process the extensions
for ext in selectedExtensions:
module = extensions.extDict[ext]
#TODO: figure out a way to do this without crashing...
#tempRoutines = copy.deepcopy(myRoutines)
module.run_extension(myRoutines, outroot, extensionOptions)
if verbose:
print "Total time taken: %.2f seconds\n" % (time.time() - start)
return header, lines, missedligandresidues
def runPDB2PQR(pdblist, ff,
outname = "",
ph = None,
verbose = False,
selectedExtensions = [],
extensionOptions = utilities.ExtraOptions(),
propkaOptions = None,
clean = False,
neutraln = False,
neutralc = False,
ligand = None,
assign_only = False,
chain = False,
debump = True,
opt = True,
typemap = False,
userff = None,
usernames = None,
ffout = None,
commandLine=None,
include_old_header=False):
"""
Run the PDB2PQR Suite
Arguments:
pdblist: The list of objects that was read from the PDB file
given as input (list)
ff: The name of the forcefield (string)
Keyword Arguments:
outname: The name of the desired output file
ph: The desired ph of the system (float)
verbose: When True, script will print information to stdout
When False, no detailed information will be printed (float)
extensions: List of extensions to run
extensionOptions:optionParser like option object that is passed to each object.
propkaOptions:optionParser like option object for propka30.
clean: only return original PDB file in aligned format.
neutraln: Make the N-terminus of this protein neutral
neutralc: Make the C-terminus of this protein neutral
ligand: Calculate the parameters for the ligand in mol2 format at the given path.
assign_only: Only assign charges and radii - do not add atoms, debump, or optimize.
chain: Keep the chain ID in the output PQR file
debump: When 1, debump heavy atoms (int)
opt: When 1, run hydrogen optimization (int)
typemap: Create Typemap output.
userff: The user created forcefield file to use. Overrides ff.
usernames: The user created names file to use. Required if using userff.
ffout: Instead of using the standard canonical naming scheme for residue and atom names, +
use the names from the given forcefield
commandLine: command line used (if any) to launch the program. Included in output header.
include_old_header: Include most of the PDB header in output.
Returns
header: The PQR file header (string)
lines: The PQR file atoms (list)
missedligandresidues: A list of ligand residue names whose charges could
not be assigned (ligand)
"""
pkaname = ""
outroot = ""
lines = []
Lig = None
atomcount = 0 # Count the number of ATOM records in pdb
outroot = utilities.getPQRBaseFileName(outname)
if not ph is None:
pka = True
pkaname = outroot + ".propka"
#TODO: What? Shouldn't it be up to propka on how to handle this?
if os.path.isfile(pkaname):
os.remove(pkaname)
else:
pka = False
start = time.time()
if verbose:
print "Beginning PDB2PQR...\n"
myDefinition = Definition()
if verbose:
print "Parsed Amino Acid definition file."
# Check for the presence of a ligand! This code is taken from pdb2pka/pka.py
if not ligand is None:
from pdb2pka.ligandclean import ligff
myProtein, myDefinition, Lig = ligff.initialize(myDefinition, ligand, pdblist, verbose)
for atom in myProtein.getAtoms():
if atom.type == "ATOM":
atomcount += 1
else:
myProtein = Protein(pdblist, myDefinition)
if verbose:
print "Created protein object -"
print "\tNumber of residues in protein: %s" % myProtein.numResidues()
print "\tNumber of atoms in protein : %s" % myProtein.numAtoms()
myRoutines = Routines(myProtein, verbose)
for residue in myProtein.getResidues():
multoccupancy = 0
for atom in residue.getAtoms():
if atom.altLoc != "":
multoccupancy = 1
txt = "Warning: multiple occupancies found: %s in %s\n" % (atom.name, residue)
sys.stderr.write(txt)
if multoccupancy == 1:
myRoutines.warnings.append("WARNING: multiple occupancies found in %s,\n" % (residue))
myRoutines.warnings.append(" at least one of the instances is being ignored.\n")
myRoutines.setTermini(neutraln, neutralc)
myRoutines.updateBonds()
if clean:
header = ""
lines = myProtein.printAtoms(myProtein.getAtoms(), chain)
# Process the extensions
for ext in selectedExtensions:
module = extensions.extDict[ext]
tempRoutines = copy.deepcopy(myRoutines)
module.run_extension(tempRoutines, outroot, extensionOptions)
if verbose:
print "Total time taken: %.2f seconds\n" % (time.time() - start)
#Be sure to include None for missed ligand residues
return header, lines, None
#remove any future need to convert to lower case
if not ff is None:
ff = ff.lower()
if not ffout is None:
ffout = ffout.lower()
if not assign_only:
# It is OK to process ligands with no ATOM records in the pdb
if atomcount == 0 and Lig != None:
pass
else:
myRoutines.findMissingHeavy()
myRoutines.updateSSbridges()
if debump:
myRoutines.debumpProtein()
if pka:
myRoutines.runPROPKA(ph, ff, outroot, pkaname, propkaOptions)
myRoutines.addHydrogens()
myhydRoutines = hydrogenRoutines(myRoutines)
if debump:
myRoutines.debumpProtein()
if opt:
myhydRoutines.setOptimizeableHydrogens()
myhydRoutines.initializeFullOptimization()
myhydRoutines.optimizeHydrogens()
else:
myhydRoutines.initializeWaterOptimization()
myhydRoutines.optimizeHydrogens()
# Special for GLH/ASH, since both conformations were added
myhydRoutines.cleanup()
else: # Special case for HIS if using assign-only
for residue in myProtein.getResidues():
if isinstance(residue, HIS):
myRoutines.applyPatch("HIP", residue)
myRoutines.setStates()
myForcefield = Forcefield(ff, myDefinition, userff, usernames)
hitlist, misslist = myRoutines.applyForcefield(myForcefield)
ligsuccess = 0
if not ligand is None:
# If this is independent, we can assign charges and radii here
for residue in myProtein.getResidues():
if isinstance(residue, LIG):
templist = []
Lig.make_up2date(residue)
for atom in residue.getAtoms():
atom.ffcharge = Lig.ligand_props[atom.name]["charge"]
atom.radius = Lig.ligand_props[atom.name]["radius"]
if atom in misslist:
misslist.pop(misslist.index(atom))
templist.append(atom)
charge = residue.getCharge()
if abs(charge - int(charge)) > 0.001:
# Ligand parameterization failed
myRoutines.warnings.append("WARNING: PDB2PQR could not successfully parameterize\n")
myRoutines.warnings.append(" the desired ligand; it has been left out of\n")
myRoutines.warnings.append(" the PQR file.\n")
myRoutines.warnings.append("\n")
# remove the ligand
myProtein.residues.remove(residue)
for myChain in myProtein.chains:
if residue in myChain.residues: myChain.residues.remove(residue)
else:
ligsuccess = 1
# Mark these atoms as hits
hitlist = hitlist + templist
# Temporary fix; if ligand was successful, pull all ligands from misslist
if ligsuccess:
templist = misslist[:]
for atom in templist:
if isinstance(atom.residue, (Amino, Nucleic)):
continue
misslist.remove(atom)
# Create the Typemap
if typemap:
typemapname = "%s-typemap.html" % outroot
myProtein.createHTMLTypeMap(myDefinition, typemapname)
# Grab the protein charge
reslist, charge = myProtein.getCharge()
# If we want a different naming scheme, use that
if not ffout is None:
scheme = ffout
userff = None # Currently not supported
if scheme != ff:
myNameScheme = Forcefield(scheme, myDefinition, userff)
else:
myNameScheme = myForcefield
myRoutines.applyNameScheme(myNameScheme)
header = printPQRHeader(pdblist, misslist, reslist, charge, ff,
myRoutines.getWarnings(), ph, ffout, commandLine,
include_old_header=include_old_header)
lines = myProtein.printAtoms(hitlist, chain)
# Determine if any of the atoms in misslist were ligands
missedligandresidues = []
for atom in misslist:
if isinstance(atom.residue, (Amino, Nucleic)):
continue
if atom.resName not in missedligandresidues:
missedligandresidues.append(atom.resName)
# Process the extensions
for ext in selectedExtensions:
module = extensions.extDict[ext]
#TODO: figure out a way to do this without crashing...
#tempRoutines = copy.deepcopy(myRoutines)
module.run_extension(myRoutines, outroot, extensionOptions)
if verbose:
print "Total time taken: %.2f seconds\n" % (time.time() - start)
return header, lines, missedligandresidues